Government Lies, Corruption and Mismanagement
NATIONAL LABOR RELATIONS BOARD v. NOEL CANNING ET AL.
On Thursday, the U.S. Supreme Court released its decision in National Labor Relations Board v. Noel Canning, a case that piqued the interest of many constitutional scholars by questioning whether the president can legally appoint government officials without Senate approval during congressional recesses—even when the Senate is meeting in “pro forma” sessions where a few senators gavel in for only a few minutes. The answer to the question, according to all nine justices, is “no.” Although the decision’s impact on executive appointments will be significant, it will also have practical impacts on the National Labor Relations Board—and by extension the labor and employment laws affecting U.S. employers—because the appointments in question were for three NLRB members who issued decisions for 18 months.
U.S. Supreme Court
NATIONAL LABOR RELATIONS BOARD v. NOEL CANNING ET AL.
June 26, 2014
ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE DISTRICT OF COLUMBIA CIRCUIT
(Bench Opinion)OCTOBER TERM, 20131
No. 12-1281. Argued January 13, 2014-Decided June 26, 2014
Respondent Noel Canning, a Pepsi-Cola distributor, asked the D. C. Circuit to set aside an order of the National Labor Relations Board, claiming that the Board lacked a quorum because three of the five Board members had been invalidly appointed. The nominations of the three members in question were pending in the Senate when it passed a December 17, 2011, resolution providing for a series of "pro forma sessions," with "no business . . . transacted," every Tuesday and Friday through January 20, 2012. S. J., 112th Cong., 1st Sess., 923. Invoking the Recess Appointments Clause-which gives the President the power "to fill up all Vacancies that may happen during the Recess of the Senate," Art. II, §2, cl. 3-the President appointed the three members in question between the January 3 and January 6 pro forma sessions. Noel Canning argued primarily that the appointments were invalid because the 3-day adjournment between those two sessions was not long enough to trigger the Recess Appointments Clause. The D. C. Circuit agreed that the appointments fell outside the scope of the Clause, but on different grounds. It held that the phrase "the recess," as used in the Clause, does not include intra-session recesses, and that the phrase "vacancies that may happen during the recess" applies only to vacancies that first come into existence during a recess.
1. The Recess Appointments Clause empowers the President to fill any existing vacancy during any recess-intra-session or inter-session-of sufficient length. Pp. 5-33.
(a) Two background considerations are relevant to the questions here. First, the Recess Appointments Clause is a subsidiary method for appointing officers of the United States. The Founders intended the norm to be the method of appointment in Article II, §2, cl. 2, which requires Senate approval of Presidential nominations, at least for principal officers. The Recess Appointments Clause reflects the tension between the President's continuous need for "the assistance of subordinates," Myers v. United States, 272 U.S. 52 , 117, and the Senate's early practice of meeting for a single brief session each year. The Clause should be interpreted as granting the President the power to make appointments during a recess but not offering the President the authority routinely to avoid the need for Senate confirmation.
Second, in interpreting the Clause, the Court puts significant weight upon historical practice. The longstanding "practice of the government," McCulloch v. Maryland, 4 Wheat. 316, 401, can inform this Court's determination of "what the law is" in a separation-of-powers case, Marbury v. Madison, 1 Cranch 137, 176. See also, e.g., Mistretta v. United States, 488 U.S. 361 , 401; The Pocket Veto Case, 279 U.S. 655 , 689-690. There is a great deal of history to consider here, for Presidents have made recess appointments since the beginning of the Republic. Their frequency suggests that the Senate and President (*2) have recognized that such appointments can be both necessary and appropriate in certain circumstances. The Court, in interpreting the Clause for the first time, must hesitate to upset the compromises and working arrangements that the elected branches of Government themselves have reached. Pp. 5-9.
(b) The phrase "the recess of the Senate" applies to both inter-session recess (i.e., breaks between formal sessions of the Senate) and intra-session recesses (i.e., breaks in the midst of a formal session) of substantial length. The constitutional text is ambiguous. Founding-era dictionaries and usages show that the phrase "the recess" can encompass intra-session breaks. And this broader interpretation is demanded by the purpose of the Clause, which is to allow the President to make appointments so as to ensure the continued functioning of the Government while the Senate is away. The Senate is equally away and unavailable to participate in the appointments process during both an inter-session and an intra-session recess. History offers further support for this interpretation. From the founding until the Great Depression, every time the Senate took a substantial, non-holiday intra-session recess, the President made recess appointments. President Andrew Johnson made the first documented intra-session recess appointments in 1867 and 1868, and Presidents made similar appointments in 1921 and 1929. Since 1929, and particularly since the end of World War II, Congress has shortened its inter-session breaks and taken longer and more frequent intra-session breaks; Presidents accordingly have made more intra-session recess appointments. Meanwhile, the Senate has never taken any formal action to deny the validity of intra-session recess appointments. In 1905, the Senate Judiciary Committee defined "the recess" as "the period of time when the Senate" is absent and cannot "participate as a body in making appointments," S. Rep. No. 4389, 58th Cong., 3d Sess., p. 2, and that functional definition encompasses both intra-session and inter-session recesses. A 1940 law regulating the payment of recess appointees has also been interpreted functionally by the Comptroller General (an officer of the Legislative Branch). In sum, Presidents have made intra-session recess appointments for a century and a half, and the Senate has never taken formal action to oppose them. That practice is long enough to entitle it to "great weight in a proper interpretation" of the constitutional provision. The Pocket Veto Case, supra, at 689.
The Clause does not say how long a recess must be in order to fall within the Clause, but even the Solicitor General concedes that a 3-day recess would be too short. The Adjournments Clause, Art. I, §5, cl. 4, reflects the fact that a 3-day break is not a significant interruption of legislative business. A Senate recess that is so short that it does not require the consent of the House under that Clause is not long enough to trigger the President's recess-appointment power. Moreover, the Court has not found a single example of a recess appointment (*3) made during an intra-session recess that was shorter than 10 days. There are a few examples of inter-session recess appointments made during recesses of less than 10 days, but these are anomalies. In light of historical practice, a recess of more than 3 days but less than 10 days is presumptively too short to fall within the Clause. The word "presumptively" leaves open the possibility that a very unusual circumstance could demand the exercise of the recess-appointment power during a shorter break. Pp. 9-21.
(c) The phrase "vacancies that may happen during the recess of the Senate," Art. II, §2, cl. 3, applies both to vacancies that first come into existence during a recess and to vacancies that initially occur before a recess but continue to exist during the recess. Again, the text is ambiguous. As Thomas Jefferson observed, the Clause is "certainly susceptible of two constructions." Letter to Wilson Cary Nicholas (Jan. 26, 1802), in 36 Papers of Thomas Jefferson 433. It "may mean 'vacancies that may happen to be' or 'may happen to fall' " during a recess. Ibid. And, as Attorney General Wirt wrote in 1821, the broader reading is more consonant with the "reason and spirit" of the Clause. 1 Op. Atty. Gen. 632. The purpose of the Clause is to permit the President, who is always acting to execute the law, to obtain the assistance of subordinate officers while the Senate, which acts only in intervals, is unavailable to confirm them. If a vacancy arises too late in the session for the President and Senate to have an opportunity to select a replacement, the narrower reading could paralyze important functions of the Federal Government, particularly at the time of the founding. The broader interpretation ensures that offices needing to be filled can be filled. It does raise a danger that the President may attempt to use the recess-appointment power to circumvent the Senate's advice and consent role. But the narrower interpretation risks undermining constitutionally conferred powers more seriously and more often. It would prevent a President from making any recess appointment to fill a vacancy that arose before a recess, no matter who the official, how dire the need, how uncontroversial the appointment, and how late in the session the office fell vacant.
Historical practice also strongly favors the broader interpretation. The tradition of applying the Clause to pre-recess vacancies dates at least to President Madison. Nearly every Attorney General to consider the question has approved the practice, and every President since James Buchanan has made recess appointments to pre-existing vacancies. It is a fair inference from the historical data that a large proportion of recess appointments over our Nation's history have filled pre-recess vacancies. The Senate Judiciary Committee in 1863 did issue a report disagreeing with the broader interpretation, and Congress passed a law known as the Pay Act prohibiting payment of recess appointments to pre-recess vacancies soon after. However, the Senate subsequently abandoned its hostility. In 1940, (*4) the Senate amended the Pay Act to permit payment of recess appointees in circumstances that would be unconstitutional under the narrower interpretation. In short, Presidents have made recess appointments to preexisting vacancies for two centuries, and the Senate as a body has not countered this practice for nearly three-quarters of a century, perhaps longer. The Court is reluctant to upset this traditional practice where doing so would seriously shrink the authority that Presidents have believed existed and have exercised for so long. Pp. 21-33.
2. For purposes of the Recess Appointments Clause, the Senate is in session when it says that it is, provided that, under its own rules, it retains the capacity to transact Senate business.
This standard is consistent with the Constitution's broad delegation of authority to the Senate to determine how and when to conduct its business, as recognized by this Court's precedents. See Art. I, §5, cl. 2; Marshall Field & Co. v. Clark, 143 U. S. 649 , 672; United States v. Ballin, 144 U. S. 1 , 5, 9. Although the Senate's own determination of when it is and is not in session should be given great weight, the Court's deference cannot be absolute. When the Senate is without the capacity to act, under its own rules, it is not in session even if it so declares.
Under the standard set forth here, the Senate was in session during the pro forma sessions at issue. It said it was in session, and Senate rules make clear that the Senate retained the power to conduct business. The Senate could have conducted business simply by passing a unanimous consent agreement. In fact, it did so; it passed a bill by unanimous consent during its pro forma session on December 23, 2011. See 2011 S. J. 924; Pub. L. 112-78 . The Court will not, as the Solicitor General urges, engage in an in-depth factual appraisal of what the Senate actually did during its pro forma sessions in order to determine whether it was in recess or in session for purposes of the Recess Appointments Clause.
Because the Senate was in session during its pro forma sessions, the President made the recess appointments at issue during a 3-day recess. Three days is too short a time to bring a recess within the scope of the Clause, so the President lacked the authority to make those appointments. Pp. 33-41.
705 F. 3d 490 , affirmed.
BREYER, J., delivered the opinion of the Court, in which KENNEDY, GINSBURG, SOTOMAYOR, and KAGAN, JJ., joined. SCALIA, J., filed an opinion concurring in the judgment, in which ROBERTS, C. J., and THOMAS and ALITO, JJ., joined.
Opinion of the Court
JUSTICE BREYER delivered the opinion of the Court.
54NLRB v. CANNING
Ordinarily the President must obtain "the Advice and Consent of the Senate" before appointing an "Officer of the United States." U.S. Const., Art. II, §2, cl. 2 . But the Recess Appointments Clause creates an exception. It gives the President alone the power "to fill up all Vacancies that may happen during the Recess of the Senate, by granting Commissions which shall expire at the End of their next Session." Art. II, §2, cl. 3 . We here consider three questions about the application of this Clause.
The first concerns the scope of the words "recess of the Senate." Does that phrase refer only to an inter-(*5) session recess (i.e., a break between formal sessions of Congress), or does it also include an intra-session recess, such as a summer recess in the midst of a session? We conclude that the Clause applies to both kinds of recess.
The second question concerns the scope of the words "vacancies that may happen." Does that phrase refer only to vacancies that first come into existence during a recess, or does it also include vacancies that arise prior to a recess but continue to exist during the recess? We conclude that the Clause applies to both kinds of vacancy.
The third question concerns calculation of the length of a "recess." The President made the appointments here at issue on January 4, 2012. At that time the Senate was in recess pursuant to a December 17, 2011, resolution providing for a series of brief recesses punctuated by "pro forma sessions," with "no business . . . transacted," every Tuesday and Friday through January 20, 2012. S. J., 112th Cong., 1st Sess., 923 (2011) (hereinafter 2011 S. J.). In calculating the length of a recess are we to ignore the pro forma sessions, thereby treating the series of brief recesses as a single, month-long recess? We conclude that we cannot ignore these pro forma sessions.
Our answer to the third question means that, when the appointments before us took place, the Senate was in the midst of a 3-day recess. Three days is too short a time to bring a recess within the scope of the Clause. Thus we conclude that the President lacked the power to make the recess appointments here at issue.
The case before us arises out of a labor dispute. The National Labor Relations Board (NLRB) found that a Pepsi-Cola distributor, Noel Canning, had unlawfully refused to reduce to writing and execute a collective-bargaining agreement with a labor union. The Board ordered the distributor to execute the agreement and to make employees whole for any losses. Noel Canning, 358 N. L. R. B. No. 4 (2012).
The Pepsi-Cola distributor subsequently asked the Court of Appeals for the District of Columbia Circuit to set the Board's order aside. It claimed that three of the five Board members had been invalidly appointed, leaving the Board without the three lawfully appointed members necessary for it to act. See 29 U.S.C. §160(f) (providing for judicial review); §153(a) (providing for a 5-member Board); §153(b) (providing for a 3-member quorum); New Process Steel, L. P. v. NLRB, 560 U.S. 674 , 687-688 (2010) (in the absence of a lawfully appointed quorum, the Board cannot exercise its powers).
The three members in question were Sharon Block, Richard Griffin, and Terence Flynn. In 2011 the President had nominated each of them to the Board. As of January 2012, Flynn's nomination had been pending in the Senate awaiting confirmation for approximately a year. The nominations of each of the other two had been pending for a few weeks. On January 4, 2012, the President, invoking the Recess Appointments Clause, appointed all three to the Board.
The distributor argued that the Recess Appointments Clause did not authorize those appointments. It pointed out that on December 17, 2011, the Senate, by unanimous consent, had adopted a resolution providing that it would (*6) take a series of brief recesses beginning the following day. See 2011 S. J. 923. Pursuant to that resolution, the Senate held pro forma sessions every Tuesday and Friday until it returned for ordinary business on January 23, 2012. Ibid.; 158 Cong. Rec. S1-S11 (Jan. 3-20, 2012). The President's January 4 appointments were made between the January 3 and January 6 pro forma sessions. In the distributor's view, each pro forma session terminated the immediately preceding recess. Accordingly, the appointments were made during a 3-day adjournment, which is not long enough to trigger the Recess Appointments Clause.
The Court of Appeals agreed that the appointments fell outside the scope of the Clause. But the court set forth different reasons. It held that the Clause's words "the recess of the Senate" do not include recesses that occur within a formal session of Congress, i.e., intra-session recesses. Rather those words apply only to recesses between those formal sessions, i.e., inter-session recesses. Since the second session of the 112th Congress began on January 3, 2012, the day before the President's appointments, those appointments occurred during an intra-session recess, and the appointments consequently fell outside the scope of the Clause. 705 F. 3d 490 , 499-507 (CADC 2013).
The Court of Appeals added that, in any event, the phrase "vacancies that may happen during the recess" applies only to vacancies that come into existence during a recess. Id., at 507-512 . The vacancies that Members Block, Griffin, and Flynn were appointed to fill had arisen before the beginning of the recess during which they were appointed. For this reason too the President's appointments were invalid. And, because the Board lacked a quorum of validly appointed members when it issued its order, the order was invalid. 29 U.S.C. §153(b) ; New Process Steel, supra .
We granted the Solicitor General's petition for certio-rari. We asked the parties to address not only the Court of Appeals' interpretation of the Clause but also the distributor's initial argument, namely, "[w]hether the President's recess-appointment power may be exercised when the Senate is convening every three days in pro forma sessions." 570 U.S. ___ (2013).
We shall answer all three questions presented. We recognize that the President has nominated others to fill the positions once occupied by Members Block, Griffin, and Flynn, and that the Senate has confirmed these successors. But, as the parties recognize, the fact that the Board now unquestionably has a quorum does not moot the controversy about the validity of the previously entered Board order. And there are pending before us petitions from decisions in other cases involving challenges to the appointment of Board Member Craig Becker. The President appointed Member Becker during an intra-session recess that was not punctuated by pro forma ses-sions, and the vacancy Becker filled had come into existence prior to the recess. See Congressional Research Service, H. Hogue, M. Carey, M. Greene, & M. Bearden, The Noel Canning Decision and Recess Appointments Made from 1981-2013, p. 28 (Feb. 4, 2013) (hereinafter (*7) The Noel Canning Decision); NLRB, Members of the NLRB since 1935, online at http://www.nlrb.gov/ who-we-are/board/members-nlrb-1935 (all Internet materials as visited June 24, 2014, and available in Clerk of Court's case file). Other cases involving similar challenges are also pending in the Courts of Appeals. E.g., NLRB v. New Vista Nursing & Rehabilitation, No. 11-3440 etc. (CA3). Thus, we believe it is important to answer all three questions that this case presents.
Before turning to the specific questions presented, we shall mention two background considerations that we find relevant to all three. First, the Recess Appointments Clause sets forth a subsidiary, not a primary, method for appointing officers of the United States. The immediately preceding Clause-Article II, Section 2, Clause 2 -provides the primary method of appointment. It says that the President "shall nominate, and by and with the Advice and Consent of the Senate, shall appoint Ambassadors, other public Ministers and Consuls, Judges of the supreme Court, and all other Officers of the United States" (emphasis added).
The Federalist Papers make clear that the Founders intended this method of appointment, requiring Senate approval, to be the norm (at least for principal officers). Alexander Hamilton wrote that the Constitution vests the power of nomination in the President alone because "one man of discernment is better fitted to analise and estimate the peculiar qualities adapted to particular offices, than a body of men of equal, or perhaps even of superior dis-cernment." The Federalist No. 76, p. 510 (J. Cooke ed. 1961). At the same time, the need to secure Senate approval provides "an excellent check upon a spirit of favoritism in the President, and would tend greatly to preventing the appointment of unfit characters from State prejudice, from family connection, from personal attachment, or from a view to popularity." Id., at 513. Hamilton further explained that the
"ordinary power of appointment is confided to the President and Senate jointly, and can therefore only be exercised during the session of the Senate; but as it would have been improper to oblige this body to be continually in session for the appointment of officers; and as vacancies might happen in their recess, which it might be necessary for the public service to fill without delay, the succeeding clause is evidently intended to authorise the President singly to make temporary appointments." Id., No. 67, at 455.
Thus the Recess Appointments Clause reflects the tension between, on the one hand, the President's continuous need for "the assistance of subordinates," Myers v. United States, 272 U.S. 52 , 117 (1926), and, on the other, the Senate's practice, particularly during the Republic's early years, of meeting for a single brief session each year, see Art. I, §4, cl. 2 ; Amdt. 20, §2 (requiring the Senate to "assemble" only "once in every year"); 3 J. Story, Commentaries on the Constitution of the United States §1551, p. 410 (1833) (it would be "burthensome to the senate, and expensive to the public" to require the Senate to be "perpetually in session"). We seek to interpret the Clause as granting the President the power to make appointments during (*8) a recess but not offering the President the author-ity routinely to avoid the need for Senate confirmation.
Second, in interpreting the Clause, we put significant weight upon historical practice. For one thing, the inter-pretive questions before us concern the allocation of power between two elected branches of Government. Long ago Chief Justice Marshall wrote that
"a doubtful question, one on which human reason may pause, and the human judgment be suspended, in the decision of which the great principles of liberty are not concerned, but the respective powers of those who are equally the representatives of the people, are to be adjusted; if not put at rest by the practice of the government, ought to receive a considerable impression from that practice." McCulloch v. Maryland, 4 Wheat. 316 , 401 (1819).
And we later confirmed that "long settled and established practice is a consideration of great weight in a proper interpretation of constitutional provisions" regulating the relationship between Congress and the President. The Pocket Veto Case, 279 U.S. 655 , 689 (1929); see also id., at 690 "A practice of at least twenty years duration 'on the part of the executive department, acquiesced in by the legislative department, . . . is entitled to great regard in determining the true construction of a constitutional provision the phraseology of which is in any respect of doubtful meaning' " (quoting State v. South Norwalk, 77 Conn. 257 , 264 , 58 A. 759, 761 (1904))).
We recognize, of course, that the separation of powers can serve to safeguard individual liberty, Clinton v. City of New York, 524 U.S. 417 , 449-450 (1998) (KENNEDY, J., concurring), and that it is the "duty of the judicial department"-in a separation-of-powers case as in any other-"to say what the law is," Marbury v. Madison, 1 Cranch 137 , 177 (1803). But it is equally true that the longstanding "practice of the government," McCulloch, supra, at 401 , can inform our determination of "what the law is," Marbury, supra, at 177 .
That principle is neither new nor controversial. As James Madison wrote, it "was foreseen at the birth of the Constitution, that difficulties and differences of opinion might occasionally arise in expounding terms & phrases necessarily used in such a charter . . . and that it might require a regular course of practice to liquidate & settle the meaning of some of them." Letter to Spencer Roane (Sept. 2, 1819), in 8 Writings of James Madison 450 (G. Hunt ed. 1908). And our cases have continually confirmed Madison's view. E.g., Mistretta v. United States, 488 U.S. 361 , 401 (1989); Dames & Moore v. Regan, 453 U.S. 654 , 686 (1981); Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579 , 610-611 (1952) (Frankfurter, J., concurring); The Pocket Veto Case, supra, at 689-690 ; Ex parte Grossman, 267 U.S. 87 , 118-119 (1925); United States v. Mid west Oil Co., 236 U.S. 459 , 472-474 (1915); McPherson v. Blacker, 146 U.S. 1 , 27 (1892); McCulloch, supra ; Stuart v. Laird, 1 Cranch 299 (1803).
These precedents show that this Court has treated practice as an important interpretive factor even when the nature or longevity of that practice is subject to dispute, and even when that practice began after the founding era. See Mistretta, supra, 400-401 ("While these [practices] spawned spirited discussion and frequent criticism, . . . 'traditional ways of conducting government . . . give meaning' to the Constitution" (quoting Youngstown, supra, at 610 ) (Frankfurter, J., concurring)); Regan, supra, at 684 ("Even if the pre-1952 (practice) should be disregarded, (*9) congressional acquiescence in (a practice) since that time supports the President's power to act here"); The Pocket Veto Case, supra, at 689-690 (postfounding practice is entitled to "great weight"); Grossman, supra, at 118-119 (postfounding practice "strongly sustains" a "construction" of the Constitution).
There is a great deal of history to consider here. Presidents have made recess appointments since the beginning of the Republic. Their frequency suggests that the Senate and President have recognized that recess appointments can be both necessary and appropriate in certain circumstances. We have not previously interpreted the Clause, and, when doing so for the first time in more than 200 years, we must hesitate to upset the compromises and working arrangements that the elected branches of Government themselves have reached.
The first question concerns the scope of the phrase "the recess of the Senate." Art. II, §2, cl. 3 (emphasis added). The Constitution provides for congressional elections every two years. And the 2-year life of each elected Congress typically consists of two formal 1-year sessions, each separated from the next by an "inter-session recess." Congressional Research Service, H. Hogue, Recess Appointments: Frequently Asked Questions 2 (2013). The Senate or the House of Representatives announces an inter-session recess by approving a resolution stating that it will "adjourn sine die," i.e., without specifying a date to return (in which case Congress will reconvene when the next formal session is scheduled to begin).
The Senate and the House also take breaks in the midst of a session. The Senate or the House announces any such "intra-session recess" by adopting a resolution stating that it will "adjourn" to a fixed date, a few days or weeks or even months later. All agree that the phrase "the recess of the Senate" covers inter-session recesses. The question is whether it includes intra-session recesses as well.
In our view, the phrase "the recess" includes an intra-session recess of substantial length. Its words taken literally can refer to both types of recess. Founding-era dictionaries define the word "recess," much as we do today, simply as "a period of cessation from usual work." 13 The Oxford English Dictionary 322-323 (2d ed. 1989) (hereinafter OED) (citing 18th- and 19th-century sources for that definition of "recess"); 2 N. Webster, An American Dictionary of the English Language (1828) ("[r]emission or suspension of business or procedure"); 2 S. Johnson, A Dictionary of the English Language 1602-1603 (4th ed. 1773) (hereinafter Johnson) (same). The Founders themselves used the word to refer to intra-session, as well as to inter-session, breaks. See, e.g., 3 Records of the Federal Convention of 1787, p. 76 (M. Farrand rev. 1966) (hereinafter Farrand) (letter from George Washington to John Jay using "the recess" to refer to an intra-session break of the Constitutional Convention); id., at 191 (speech of Luther Martin with a similar usage); 1 T. Jefferson, A Manual of Parliamentary Practice §LI, p. 165 (2d ed. 1812) (describing a "recess by adjournment" which did not end a session).(*10)
We recognize that the word "the" in "the recess" might suggest that the phrase refers to the single break separating formal sessions of Congress. That is because the word "the" frequently (but not always) indicates "a particular thing." 2 Johnson 2003. But the word can also refer "to a term used generically or universally." 17 OED 879. The Constitution, for example, directs the Senate to choose a President pro tempore "in the Absence of the Vice-President." Art. I, §3, cl. 5 (emphasis added). And the Federalist Papers refer to the chief magistrate of an ancient Achaean league who "administered the government in the recess of the Senate." The Federalist No. 18, at 113 (J. Madison) (emphasis added). Reading "the" generically in this way, there is no linguistic problem applying the Clause's phrase to both kinds of recess. And, in fact, the phrase "the recess" was used to refer to intra-session recesses at the time of the founding. See, e.g., 3 Farrand 76 (letter from Washington to Jay); New Jersey Legislative-Council Journal, 5th Sess., 1st Sitting 70, 2d Sitting 9 (1781) (twice referring to a 4-month, intra-session break as "the Recess"); see also Brief for Petitioner 14-16 (listing examples).
The constitutional text is thus ambiguous. And we believe the Clause's purpose demands the broader interpretation. The Clause gives the President authority to make appointments during "the recess of the Senate" so that the President can ensure the continued functioning of the Federal Government when the Senate is away. The Senate is equally away during both an inter-session and an intra-session recess, and its capacity to participate in the appointments process has nothing to do with the words it uses to signal its departure.
History also offers strong support for the broad interpretation. We concede that pre-Civil War history is not helpful. But it shows only that Congress generally took long breaks between sessions, while taking no significant intra-session breaks at all (five times it took a break of a week or so at Christmas). See Appendix A, infra. Obviously, if there are no significant intra-session recesses, there will be no intra-session recess appointments. In 1867 and 1868, Congress for the first time took substantial, nonholiday intra-session breaks, and President Andrew Johnson made dozens of recess appointments. The Federal Court of Claims upheld one of those specific appointments, writing "we have no doubt that a vacancy occurring while the Senate was thus temporarily adjourned" during the "first session of the Fortieth Congress" was "legally filled by appointment of the President alone." Gould v. United States, 19 Ct. Cl. 593, 595-596 (1884) (emphasis added). Attorney General Evarts also issued three opinions concerning the constitutionality of President Johnson's appointments, and it apparently did not occur to him that the distinction between intra-session and inter-session recesses was significant. See 12 Op. Atty. Gen. 449 (1868); 12 Op. Atty. Gen. 455 (1868); 12 Op. Atty. Gen. 469 (1868). Similarly, though the 40th Congress impeached President Johnson on charges relating to his appointment power, he was not (*11) accused of violating the Constitution by making intra-session recess appointments. Hartnett, Recess Appointments of Article III Judges: Three Constitutional Questions, 26 Cardozo L. Rev. 377, 409 (2005).
In all, between the founding and the Great Depression, Congress took substantial intra-session breaks (other than holiday breaks) in four years: 1867, 1868, 1921, and 1929. Appendix A, infra. And in each of those years the President made intra-session recess appointments. See App. to Brief for Petitioner 1a-11a.
Since 1929, and particularly since the end of World War II, Congress has shortened its inter-session breaks as it has taken longer and more frequent intra-session breaks; Presidents have correspondingly made more intra-session recess appointments. Indeed, if we include military appointments, Presidents have made thousands of intra-session recess appointments. Id., at 11a-64a. President Franklin Roosevelt, for example, commissioned Dwight Eisenhower as a permanent Major General during an intra-session recess; President Truman made Dean Acheson Under Secretary of State; and President George H. W. Bush reappointed Alan Greenspan as Chairman of the Federal Reserve Board. Id., at 11a, 12a, 40a. JUSTICE SCALIA does not dispute any of these facts.
Not surprisingly, the publicly available opinions of Presidential legal advisers that we have found are nearly unanimous in determining that the Clause authorizes these appointments. In 1921, for example, Attorney General Daugherty advised President Harding that he could make intra-session recess appointments. He reasoned:
"If the President's power of appointment is to be defeated because the Senate takes an adjournment to a specified date, the painful and inevitable result will be measurably to prevent the exercise of governmental functions. I can not bring myself to believe that the framers of the Constitution ever intended such a catastrophe to happen." 33 Op. Atty. Gen. 20, 23.
We have found memoranda offering similar advice to President Eisenhower and to every President from Carter to the present. See 36 Opinion of Office of Legal Counsel (Op. OLC) ___, ___ (2012), online at www.justice.gov/ olc/opiniondocslpro-forma-sessions-opinion.pdf; 25 Op. OLC 182 (2001); 20 Op. OLC 124, 161 (1996); 16 Op. OLC 15 (1992); 13 Op. OLC 271 (1989); 6 Op. OLC 585, 586 (1982); 3 Op. OLC 314, 316 (1979); 41 Op. Atty. Gen. 463, 466 (1960).
We must note one contrary opinion authored by President Theodore Roosevelt's Attorney General Philander Knox. Knox advised the President that the Clause did not cover a 19-day intra-session Christmas recess. 23 Op. Atty. Gen. 599 (1901). But in doing so he relied heavily upon the use of the word "the," a linguistic point that we do not find determinative. See supra, at 10. And Knox all but confessed that his interpretation ran contrary to the basic purpose of the Clause. For it would permit the Senate to adjourn for "several months," to a fixed date, and thereby "seriously curtail the President's power of making recess appointments." 23 Op. Atty. Gen., at 603. Moreover, only three days before Knox gave his opinion, the Solicitor (*12) of the Treasury came to the opposite conclusion. Reply Brief 7, n. 5. We therefore do not think Knox's isolated opinion can disturb the consensus advice within the Executive Branch taking the opposite position.
What about the Senate? Since Presidents began making intra-session recess appointments, individual Senators have taken differing views about the proper definition of "the recess." See, e.g., 130 Cong. Rec. 23234 (1984) (resolution introduced by Senator Byrd urging limits on the length of applicable intra-session recesses); Brief for Sen. Mitch McConnell et al. as Amici Curiae 26 (an intra-session adjournment does not count as "the recess"); Brief for Sen. Edward M. Kennedy as Amicus Curiae in Franklin v. United States, O. T. 2004, No. 04-5858, p. 5 (same). But neither the Senate considered as a body nor its committees, despite opportunities to express opposition to the practice of intra-session recess appointments, has done so. Rather, to the extent that the Senate or a Senate committee has expressed a view, that view has favored a functional definition of "recess," and a functional definition encompasses intra-session recesses.
Most notably, in 1905 the Senate Committee on the Judiciary objected strongly to President Theodore Roosevelt's use of the Clause to make more than 160 recess appointments during a "fictitious" inter-session recess. S. Rep. No. 4389, 58th Cong., 3d Sess., p. 2 (hereinafter 1905 Senate Report). At noon on December 7, 1903, the Senate President pro tempore had "declared" a formal, "extraordinary session" of the Senate "adjourned without day," and the next formal Senate session began immediately afterwards. 37 Cong. Rec. 544 (1903). President Roosevelt made over 160 recess appointments during the instantaneous inter-session interval. The Judiciary Committee, when stating its strong objection, defined "recess" in functional terms as
"the period of time when the Senate is not sitting in regular or extraordinary session as a branch of the Congress . . . ; when its members owe no duty of attendance; when its Chamber is empty; when, because of its absence, it can not receive communications from the President or participate as a body in making appointments." 1905 Senate Report, at 2 (emphasis deleted).
That functional definition encompasses intra-session, as well as inter-session, recesses. JUSTICE SCALIA is right that the 1905 Report did not specifically address the dis-tinction between inter-session and intra-session recesses. But the animating principle of the Report-that "recess" should be practically construed to mean a time when the Senate is unavailable to participate in the appointments process-is inconsistent with the formalistic approach that JUSTICE SCALIA endorses.
Similarly, in 1940 the Senate helped to enact a law regulating the payment of recess appointees, and the Comptroller General of the United States has interpreted that law functionally. An earlier 1863 statute had denied pay to individuals appointed to fill up vacancies first arising prior to the beginning of a recess. The Senate Judiciary Committee then believed that those vacancies (*13) fell outside the scope of the Clause. See infra, at 30. In 1940, however, the Senate amended the law to permit many of those recess appointees to be paid. Act of July 11, 54 Stat. 751 . Interpreting the amendments in 1948, the Comptroller General-who, unlike the Attorney General, is an "officer of the Legislative Branch," Bowsher v. Synar, 478 U.S. 714 , 731 (1986)-wrote:
"I think it is clear that (the Pay Act amendments) primary purpose was to relieve 'recess appointees' of the burden of serving without compensation during periods when the Senate is not actually sitting and is not available to give its advice and consent in respect to the appointment, irrespective of whether the recess of the Senate is attributable to a final adjournment sine die or to an adjournment to a specified date." 28 Comp. Gen. 30, 37.
We recognize that the Senate cannot easily register opposition as a body to every governmental action that many, perhaps most, Senators oppose. But the Senate has not been silent or passive regarding the meaning of the Clause: A Senate Committee did register opposition to President Theodore Roosevelt's use of the Clause, and the Senate as a whole has legislated in an effort to discourage certain kinds of recess appointments. And yet we are not aware of any formal action it has taken to call into question the broad and functional definition of "recess" first set out in the 1905 Senate Report and followed by the Executive Branch since at least 1921. Nor has JUSTICE SCALIA identified any. All the while, the President has made countless recess appointments during intra-session recesses.
The upshot is that restricting the Clause to inter-session recesses would frustrate its purpose. It would make the President's recess-appointment power dependent on a formalistic distinction of Senate procedure. Moreover, the President has consistently and frequently interpreted the word "recess" to apply to intra-session recesses, and has acted on that interpretation. The Senate as a body has done nothing to deny the validity of this practice for at least three-quarters of a century. And three-quarters of a century of settled practice is long enough to entitle a practice to "great weight in a proper interpretation" of the constitutional provision. The Pocket Veto Case, 279 U.S., at 689 .
We are aware of, but we are not persuaded by, three important arguments to the contrary. First, some argue that the Founders would likely have intended the Clause to apply only to inter-session recesses, for they hardly knew any other. See, e.g., Brief for Originalist Scholars as Amici Curiae 27-29. Indeed, from the founding until the Civil War inter-session recesses were the only kind of significant recesses that Congress took. The problem with this argument, however, is that it does not fully describe the relevant founding intent. The question is not: Did the Founders at the time think about intra-session recesses? Perhaps they did not. The question is: Did the Founders intend to restrict the scope of the Clause to the form of congressional recess then prevalent, or did they intend a broader scope permitting the Clause (*14) to apply, where appropriate, to somewhat changed circumstances? The Founders knew they were writing a document designed to apply to ever-changing circumstances over centuries. After all, a Constitution is "intended to endure for ages to come," and must adapt itself to a future that can only be "seen dimly," if at all. McCulloch, 4 Wheat., at 415 . We therefore think the Framers likely did intend the Clause to apply to a new circumstance that so clearly falls within its essential purposes, where doing so is consistent with the Clause's language.
Second, some argue that the intra-session interpretation permits the President to make "illogically" long recess appointments. Brief for Respondent Noel Canning 13; post, at 10 (SCALIA, J., concurring in judgment). A recess appointment made between Congress' annual sessions would permit the appointee to serve for about a year, i.e., until the "end" of the "next" Senate "session." Art. II, §2, cl. 3 . But an intra-session appointment made at the beginning or in the middle of a formal session could permit the appointee to serve for 1½ or almost 2 years (until the end of the following formal session).
We agree that the intra-session interpretation permits somewhat longer recess appointments, but we do not agree that this consequence is "illogical." A President who makes a recess appointment will often also seek to make a regular appointment, nominating the appointee and securing ordinary Senate confirmation. And the Clause ensures that the President and Senate always have at least a full session to go through the nomination and confirmation process. That process may take several months. See O'Connell, Vacant Offices: Delays in Staffing Top Agency Positions, 82 S. Cal. L. Rev. 913, 967 (2009) (from 1987 to 2005 the nomination and confirmation process took an average of 236 days for noncabinet agency heads). A recess appointment that lasts somewhat longer than a year will ensure the President the continued assistance of subordinates that the Clause permits him to obtain while he and the Senate select a regular appointee. An appointment should last until the Senate has "an opportunity to act on the subject," Story, §1551, at 410, and the Clause embodies a determination that a full session is needed to select and vet a replacement.
Third, the Court of Appeals believed that application of the Clause to intra-session recesses would introduce "vagueness" into a Clause that was otherwise clear. 705 F. 3d, at 504 . One can find problems of uncertainty, however, either way. In 1867, for example, President Andrew Johnson called a special session of Congress, which took place during a lengthy intra-session recess. Consider the period of time that fell just after the conclusion of that special session. Did that period remain an intra-session recess, or did it become an inter-session recess? Historians disagree about the answer. Compare Hartnett, 26 Cardozo L. Rev., at 408-409, with Brief for Constitutional Law Scholars as Amici Curiae 23-24.
Or suppose that Congress adjourns sine die, but it does so conditionally, so that the leadership can call (*15) the members back into session when "the public interest shall warrant it." E.g., 155 Cong. Rec. 33429 (2009); 152 Cong. Rec. 23731-23732 (2006); 150 Cong. Rec. 25925-25926 (2004). If the Senate Majority Leader were to reconvene the Senate, how would we characterize the preceding recess? Is it still inter-session? On the narrower interpretation the label matters; on the broader it does not.
The greater interpretive problem is determining how long a recess must be in order to fall within the Clause. Is a break of a week, or a day, or an hour too short to count as a "recess"? The Clause itself does not say. And JUSTICE SCALIA claims that this silence itself shows that the Framers intended the Clause to apply only to an inter-session recess. Post, at 12-13.
We disagree. For one thing, the most likely reason the Framers did not place a textual floor underneath the word "recess" is that they did not foresee the need for one. They might have expected that the Senate would meet for a single session lasting at most half a year. The Federalist No. 84, at 596 (A. Hamilton). And they might not have anticipated that intra-session recesses would become lengthier and more significant than inter-session ones. The Framers' lack of clairvoyance on that point is not dispositive. Unlike JUSTICE SCALIA, we think it most consistent with our constitutional structure to presume that the Framers would have allowed intra-session recess appointments where there was a long history of such practice.
Moreover, the lack of a textual floor raises a problem that plagues both interpretations-JUSTICE SCALIA's and ours. Today a brief inter-session recess is just as possible as a brief intra-session recess. And though JUSTICE SCALIA says that the "notion that the Constitution empowers the President to make unilateral appointments every time the Senate takes a half-hour lunch break is so absurd as to be self-refuting," he must immediately concede (in a footnote) that the President "can make recess appointments during any break between sessions, no matter how short." Post, at 11, 15, n. 4 (emphasis added).
Even the Solicitor General, arguing for a broader interpretation, acknowledges that there is a lower limit applicable to both kinds of recess. He argues that the lower limit should be three days by analogy to the Adjournments Clause of the Constitution. Tr. of Oral Arg. 11. That Clause says: "Neither House, during the Session of Congress, shall, without the Consent of the other, adjourn for more than three days." Art. I, §5, cl. 4 .
We agree with the Solicitor General that a 3-day recess would be too short. (Under Senate practice, "Sunday is generally not considered a day," and so is not counted for purposes of the Adjournments Clause. S. Doc. No. 101-28, F. Riddick & A. Frumin, Riddick's Senate Procedure: Precedents and Practices 1265 (hereinafter Riddick's).) The Adjournments Clause reflects the fact that a 3-day break is not a significant interruption of legislative business. As the Solicitor General says, it is constitutionally de minimis. Brief for Petitioner 18. A Senate recess that is so short that it does not require the consent (*16) of the House is not long enough to trigger the President's recess-appointment power.
That is not to say that the President may make recess appointments during any recess that is "more than three days." Art. I, §5, cl. 4 . The Recess Appointments Clause seeks to permit the Executive Branch to function smoothly when Congress is unavailable. And though Congress has taken short breaks for almost 200 years, and there have been many thousands of recess appointments in that time, we have not found a single example of a recess ap-pointment made during an intra-session recess that was shorter than 10 days. Nor has the Solicitor General. Reply Brief 23. Indeed, the Office of Legal Counsel once informally advised against making a recess appointment during a 6-day intra-session recess. 3 Op. OLC, at 315-316. The lack of examples suggests that the recess-appointment power is not needed in that context. (The length of a recess is "ordinarily calculated by counting the calendar days running from the day after the recess begins and including the day the recess ends." 36 Op. OLC, at ___, n. 1 (citation omitted).)
There are a few historical examples of recess appointments made during inter-session recesses shorter than 10 days. We have already discussed President Theodore Roosevelt's appointments during the instantaneous, "fictitious" recess. President Truman also made a recess appointment to the Civil Aeronautics Board during a 3-day inter-session recess. Hogue, Recess Appointments: Fre-quently Asked Questions, at 5-6. President Taft made a few appointments during a 9-day recess following his inauguration, and President Lyndon Johnson made sev-eral appointments during an 8-day recess several weeks after assuming office. Hogue, The Law: Recess Appointments to Article III Courts, 34 Presidential Studies Q. 656, 671 (2004); 106 S. Exec. J. 2 (1964); 40 S. Exec. J. 12 (1909). There may be others of which we are unaware. But when considered against 200 years of settled practice, we regard these few scattered examples as anomalies. We therefore conclude, in light of historical practice, that a recess of more than 3 days but less than 10 days is presumptively too short to fall within the Clause. We add the word "presumptively" to leave open the possibility that some very unusual circumstance-a national catastrophe, for instance, that renders the Senate unavailable but calls for an urgent response-could demand the exercise of the recess-appointment power during a shorter break. (It should go without saying-except that JUSTICE SCALIA compels us to say it-that political opposition in the Senate would not qualify as an unusual circumstance.)
In sum, we conclude that the phrase "the recess" applies to both intra-session and inter-session recesses. If a Senate recess is so short that it does not require the consent of the House, it is too short to trigger the Recess Appointments Clause. See Art. I, §5, cl. 4 . And a recess lasting less than 10 days is presumptively too short as well.
The second question concerns the scope of the phrase "vacancies that may happen during the recess of the Senate." (*17) Art. II, §2, cl. 3 (emphasis added). All agree that the phrase applies to vacancies that initially occur during a recess. But does it also apply to vacancies that initially occur before a recess and continue to exist during the recess? In our view the phrase applies to both kinds of vacancy.
We believe that the Clause's language, read literally, permits, though it does not naturally favor, our broader interpretation. We concede that the most natural meaning of "happens" as applied to a "vacancy" (at least to a modern ear) is that the vacancy "happens" when it ini-tially occurs. See 1 Johnson 913 (defining "happen" in relevant part as meaning "to fall out; to chance; to come to pass"). But that is not the only possible way to use the word.
Thomas Jefferson wrote that the Clause is "certainly susceptible of (two) constructions." Letter to Wilson Cary Nicholas (Jan. 26, 1802), in 36 Papers of Thomas Jefferson 433 (B. Oberg ed., 2009). It "may mean 'vacancies that may happen to be' or 'may happen to fall' " during a recess. Ibid. Jefferson used the phrase in the first sense when he wrote to a job seeker that a particular position was unavailable, but that he (Jefferson) was "happy that another vacancy happens wherein I can . . . avail the public of your integrity & talents," for "the office of Treasurer of the US. is vacant by the resignation of mr Meredith." Letter to Thomas Tudor Tucker (Oct. 31, 1801), in 35 id., at 530 (B. Oberg ed. 2008) (emphasis added). See also Laws Passed by the Legislature of Florida, No. 31, An Act to Organize and Regulate the Militia of the Territory of Florida §13, H. R. Exec. Doc. No. 72, 27th Cong., 3d Sess., 22 (1842) ("When any vacancy shall take place in the office of any lieutenant colonel, it shall be the duty of the colonel of the regiment in which such vacancy may happen to order an election to be held at the several precincts in the battalion in which such vacancy may happen" (emphasis added)).
Similarly, when Attorney General William Wirt advised President Monroe to follow the broader interpretation, he wrote that the "expression seems not perfectly clear. It may mean 'happen to take place:' that is, 'to originate,' " or it "may mean, also, without violence to the sense, 'happen to exist.' " 1 Op. Atty. Gen. 631, 631-632 (1823). The broader interpretation, he added, is "most accordant with" the Constitution's "reason and spirit." Id., at 632.
We can still understand this earlier use of "happen" if we think of it used together with another word that, like "vacancy," can refer to a continuing state, say, a financial crisis. A statute that gives the President authority to act in respect to "any financial crisis that may happen during his term" can easily be interpreted to include crises that arise before, and continue during, that term. Perhaps that is why the Oxford English Dictionary defines "happen" in part as "chance to be," rather than "chance to occur." 6 OED 1096 (emphasis added); see also 19 OED 383 (defining "vacancy" as the "condition of an office or post being . . . vacant").
In any event, the linguistic question here is not whether the phrase can be, but whether it must be, read more narrowly. The question is whether the Clause is ambiguous. The Pocket Veto Case (*18) , 279 U.S., at 690 . And the broader reading, we believe, is at least a permissible reading of a " 'doubtful' " phrase. Ibid. We consequently go on to consider the Clause's purpose and historical practice.
The Clause's purpose strongly supports the broader interpretation. That purpose is to permit the President to obtain the assistance of subordinate officers when the Senate, due to its recess, cannot confirm them. Attorney General Wirt clearly described how the narrower interpretation would undermine this purpose:
"Put the case of a vacancy occurring in an office, held in a distant part of the country, on the last day of the Senate's session. Before the vacancy is made known to the President, the Senate rises. The office may be an important one; the vacancy may paralyze a whole line of action in some essential branch of our internal police; the public interests may imperiously demand that it shall be immediately filled. But the vacancy happened to occur during the session of the Senate; and if the President's power is to be limited to such vacancies only as happen to occur during the recess of the Senate, the vacancy in the case put must continue, however ruinous the consequences may be to the public." 1 Op. Atty. Gen., at 632.
Examples are not difficult to imagine: An ambassadorial post falls vacant too soon before the recess begins for the President to appoint a replacement; the Senate rejects a President's nominee just before a recess, too late to select another. Wirt explained that the "substantial purpose of the constitution was to keep these offices filled," and "if the President shall not have the power to fill a vacancy thus circumstanced, . . . the substance of the constitution will be sacrificed to a dubious construction of its letter." Ibid. Thus the broader construction, encompassing vacancies that initially occur before the beginning of a recess, is the "only construction of the constitution which is compatible with its spirit, reason, and purposes; while, at the same time, it offers no violence to its language." Id., at 633.
We do not agree with JUSTICE SCALIA's suggestion that the Framers would have accepted the catastrophe envisioned by Wirt because Congress can always provide for acting officers, see 5 U.S.C. §3345 , and the President can always convene a special session of Congress, see U.S. Const., Art. II, §3 . Acting officers may have less authority than Presidential appointments. 6 Op. OLC 119, 121 (1982). Moreover, to rely on acting officers would lessen the President's ability to staff the Executive Branch with people of his own choosing, and thereby limit the President's control and political accountability. Cf. Free Enterprise Fund v. Public Company Accounting Oversight Bd., 561 U.S. 477 , 497-498 (2010). Special sessions are burdensome (and would have been especially so at the time of the founding). The point of the Recess Appointments Clause was to avoid reliance on these inadequate expedients.
At the same time, we recognize one important purpose-related consideration that argues in the opposite direction. A broad interpretation might permit a President to avoid Senate confirmations as a matter of course. If the Clause gives the (*19) President the power to "fill up all vacancies" that occur before, and continue to exist during, the Senate's recess, a President might not submit any nominations to the Senate. He might simply wait for a recess and then provide all potential nominees with recess appointments. He might thereby routinely avoid the constitutional need to obtain the Senate's "advice and consent."
Wirt thought considerations of character and politics would prevent Presidents from abusing the Clause in this way. 1 Op. Atty. Gen., at 634. He might have added that such temptations should not often arise. It is often less desirable for a President to make a recess appointment. A recess appointee only serves a limited term. That, combined with the lack of Senate approval, may diminish the recess appointee's ability, as a practical matter, to get a controversial job done. And even where the President and Senate are at odds over politically sensitive appointments, compromise is normally possible. Indeed, the 1940 Pay Act amendments represent a general compromise, for they foresee payment of salaries to recess appointees where vacancies occur before the recess began but not too long before (namely, within 30 days before). 5 U.S.C. §5503(a)(1) ; see infra, at 32. Moreover, the Senate, like the President, has institutional "resources," including political resources, "available to protect and assert its interests." Goldwater v. Carter, 444 U.S. 996 , 1004 (1979) (Rehnquist, J., concurring in judgment). In an unusual instance, where a matter is important enough to the Sen-ate, that body can remain in session, preventing recess appointments by refusing to take a recess. See Part V, infra. In any event, the Executive Branch has adhered to the broader interpretation for two centuries, and Senate confirmation has always remained the norm for officers that require it.
While we concede that both interpretations carry with them some risk of undesirable consequences, we believe the narrower interpretation risks undermining constitutionally conferred powers more seriously and more often. It would prevent the President from making any recess appointment that arose before a recess, no matter who the official, no matter how dire the need, no matter how uncontroversial the appointment, and no matter how late in the session the office fell vacant. Overall, like Attorney General Wirt, we believe the broader interpretation more consistent with the Constitution's "reason and spirit." 1 Op. Atty. Gen., at 632.
Historical practice over the past 200 years strongly favors the broader interpretation. The tradition of applying the Clause to pre-recess vacancies dates at least to President James Madison. There is no undisputed record of Presidents George Washington, John Adams, or Thomas Jefferson making such an appointment, though the Solicitor General believes he has found records showing that Presidents Washington and Jefferson did so. We know that Edmund Randolph, Washington's Attorney General, favored a narrow reading of the Clause. Randolph believed that the "Spirit of the Constitution favors the participation of the Senate (*20) in all appointments," though he did not address-let alone answer-the powerful purposive and structural arguments subsequently made by Attorney General Wirt. See Edmund Randolph's Opinion on Recess Appointments (July 7, 1792), in 24 Papers of Thomas Jefferson 166 (J. Catanzariti ed. 1990).
President Adams seemed to endorse the broader view of the Clause in writing, though we are not aware of any appointments he made in keeping with that view. See Letter to J. McHenry (Apr. 16, 1799), in 8 Works of John Adams 632-633 (C. Adams ed. 1853). His Attorney General, Charles Lee, later informed Jefferson that, in the Adams administration, "whenever an office became vacant so short a time before Congress rose, as not to give an opportunity of enquiring for a proper character, they let it lie always till recess." 36 Papers of Thomas Jefferson 433. We know that President Jefferson thought that the broad interpretation was linguistically supportable, though his actual practice is not clear. But the evidence suggests that James Madison-as familiar as anyone with the workings of the Constitutional Convention-appointed Theodore Gaillard to replace a district judge who had left office before a recess began. Hartnett, 26 Cardozo L. Rev., at 400-401. It also appears that in 1815 Madison signed a bill that created two new offices prior to a recess which he then filled later during the recess. See Act of Mar. 3, ch. 95, 3 Stat. 235 ; S. J. 13th Cong., 3d Sess., 689-690 (1815); 3 S. Exec. J. 19 (1828) (for Monday, Jan. 8, 1816). He also made recess appointments to "territorial" United States attorney and marshal positions, both of which had been created when the Senate was in session more than two years before. Act of Feb. 27, 1813, ch. 35, 2 Stat. 806 ; 3 S. Exec. J. 19. JUSTICE SCALIA refers to "written evidence of Madison's own beliefs," post, at 36, but in fact we have no direct evidence of what President Madison believed. We only know that he declined to make one appointment to a pre-recess vacancy after his Secretary of War advised him that he lacked the power. On the other hand, he did apparently make at least five other appointments to pre-recess vacancies, as JUSTICE SCALIA does not dispute.
The next President, James Monroe, received and presumably acted upon Attorney General Wirt's advice, namely that "all vacancies which, from any casualty, happen to exist at a time when the Senate cannot be consulted as to filling them, may be temporarily filled by the President." 1 Op. Atty. Gen., at 633. Nearly every subsequent Attorney General to consider the question throughout the Nation's history has thought the same. E.g., 2 Op. Atty. Gen. 525, 528 (1832); 7 Op. Atty. Gen. 186, 223 (1855); 10 Op. Atty. Gen. 356, 356-357 (1862); 12 Op. Atty. Gen. 32, 33 (1866); 12 Op. Atty. Gen., at 452; 14 Op. Atty. Gen. 562, 564 (1875); 15 Op. Atty. Gen. 207 (1877); 16 Op. Atty. Gen. 522, 524 (1880); 17 Op. Atty. Gen. 521 (1883); 18 Op. Atty. Gen. 29, 29-30 (1884); 19 Op. Atty. Gen. 261, 262 (1889); 26 Op. Atty. Gen. 234, 234-235 (1907); 30 Op. Atty. Gen. 314, 315 (1914); 41 Op. Atty. Gen. 463, 465 (1960); 3 Op. OLC 314 (1979); 6 Op. OLC 585, 586 (1982); 20 Op. OLC 124, 161 (1996); 36 Op. OLC ___ (2012). Indeed, as early as 1862, Attorney General Bates advised President Lincoln that his power to fill pre-recess vacancies [*21] was "settled . . . as far . . . as a constitutional question can be settled," 10 Op. Atty. Gen., at 356, and a century later Acting Attorney General Walsh gave President Eisenhower the same advice "without any doubt," 41 Op. Atty. Gen., at 466.
This power is important. The Congressional Research Service is "unaware of any official source of information tracking the dates of vacancies in federal offices." The Noel Canning Decision 3, n. 6. Nonetheless, we have enough information to believe that the Presidents since Madison have made many recess appointments filling vacancies that initially occurred prior to a recess. As we have just said, nearly every 19th- and 20th-century Attorney General expressing a view on the matter has agreed with William Wirt, and Presidents tend to follow the legal advice of their chief legal officers. Moreover, the Solicitor General has compiled a list of 102 (mostly uncontested) recess appointments made by Presidents going back to the founding. App. to Brief for Petitioner 65a-89a. Given the difficulty of finding accurate information about vacancy dates, that list is undoubtedly far smaller than the actual number. No one disputes that every President since James Buchanan has made recess appointments to pre-existing vacancies.
Common sense also suggests that many recess appointees filled vacancies that arose before the recess began. We have compared the list of intra-session recess appointments in the Solicitor General's brief with the chart of congressional recesses. Where a specific date of appointment can be ascertained, more than half of those intra-session appointments were made within two weeks of the beginning of a recess. That short window strongly suggests that many of the vacancies initially arose prior to the recess. See App. to Brief for Petitioner 1a-64a; Appendix A, infra. Thus, it is not surprising that the Congressional Research Service, after examining the vacancy dates associated with a random sample of 24 inter-session recess appointments since 1981, concluded that "[i]n most of the 24 cases, the preponderance of evidence indicated that the vacancy arose prior to the recess during which the appointment was made." The Noel Canning Decision 3. Further, with research assistance from the Supreme Court Library, we have examined a random sample of the recess appointments made by our two most recent Presidents, and have found that almost all of those appointments filled pre-recess vacancies: Of a sample of 21 recess appointments, 18 filled pre-recess vacancies and only 1 filled a vacancy that arose during the recess in which he was appointed. The precise date on which 2 of the vacancies arose could not be determined. See Appendix B, infra. Taken together, we think it is a fair inference that a large proportion of the recess appointments in the history of the Nation have filled pre-existing vacancies.
Did the Senate object? Early on, there was some sporadic disagreement with the broad interpretation. In 1814 Senator Gore said that if "the vacancy happen at another time, it is not the case described [*22] by the Constitution." 26 Annals of Cong. 653. In 1822 a Senate committee, while focusing on the President's power to fill a new vacancy created by statute, used language to the same effect. 38 id., at 489, 500. And early Congresses enacted statutes authorizing certain recess appointments, see post, at 31, a fact that may or may not suggest they accepted the narrower interpretation of the Clause. Most of those statutes-including the one passed by the First Congress-authorized appointments to newly created offices, and may have been addressed to the separate question of whether new offices are vacancies within the meaning of the Clause. See Letter from Alexander Hamilton to James McHenry (May 3, 1799), in 23 Papers of Alexander Hamilton 94 (H. Syrett ed. 1976) ("Vacancy is a relative term, and presupposes that the Office has been once filled"); Reply Brief 17. In any event, by 1862 Attorney General Bates could still refer to "the unbroken acquiescence of the Senate" in support of the broad interpretation. 10 Op. Atty. Gen., at 356.
Then in 1863 the Senate Judiciary Committee disagreed with the broad interpretation. It issued a report concluding that a vacancy "must have its inceptive point after one session has closed and before another session has begun." S. Rep. No. 80, 37th Cong., 3d Sess., p. 3. And the Senate then passed the Pay Act, which provided that "no money shall be paid . . . as a salary, to any person appointed during the recess of the Senate, to fill a vacancy . . . which . . . existed while the Senate was in session." Act of Feb. 9, 1863, §2, 12 Stat. 646 . Relying upon the floor statement of a single Senator, JUSTICE SCALIA suggests that the passage of the Pay Act indicates that the Senate as a whole endorsed the position in the 1863 Report. But the circumstances are more equivocal. During the floor debate on the bill, not a single Senator referred to the Report. Cong. Globe, 37th Cong., 3d Sess. 564-565 (1863). Indeed, Senator Trumbull, who introduced the Pay Act, acknowledged that there was disagreement about the underlying constitutional question. Id., at 565 ("[S]ome other persons think he has that power"). Further, if a majority of the Senate had believed appointments to pre-recess vacancies were unconstitutional, it could have attempted to do far more than temporarily dock the appointees' pay. Cf. Tenure of Office Act of 1867, §5, 14 Stat. 431 (making it a federal crime for "any person" to "accept any appointment" in certain circumstances).
In any event, the Senate subsequently abandoned its hostility. In the debate preceding the 1905 Senate Report regarding President Roosevelt's "constructive" recess appointments, Senator Tillman-who chaired the Committee that authored the 1905 Report-brought up the 1863 Report, and another Senator responded: "Whatever that report may have said in 1863, I do not think that has been the view the Senate has taken" of the issue. 38 Cong. Rec. 1606 (1904). Senator Tillman then agreed that "the Senate has acquiesced" in the President's "power to fill" pre-recess vacancies. Ibid. And Senator Tillman's 1905 Report described the Clause's purpose in terms closely echoing Attorney General Wirt. 1905 [*23] Senate Report, at 2 ("Its sole purpose was to render it certain that at all times there should be, whether the Senate was in session or not, an officer for every office" (emphasis added)).
In 1916 the Senate debated whether to pay a recess appointee who had filled a pre-recess vacancy and had not subsequently been confirmed. Both Senators to address the question-one on each side of the payment debate-agreed that the President had the constitutional power to make the appointment, and the Senate voted to pay the appointee for his service. 53 Cong. Rec. 4291-4299; 39 Stat. 818 -819. In 1927 the Comptroller General, a legislative officer, wrote that "there is no question but that the President has authority to make a recess appointment to fill any vacancy," including those that "existed while the Senate was in session." 7 Comp. Gen. 10, 11 (emphasis added). Meanwhile, Presidents continued to make appointments to pre-recess vacancies. The Solicitor General has identified 40 between 1863 and 1940, but that number is clearly not comprehensive. See, e.g., 32 Op. Atty. Gen. 271-272 (1920) (listing 5 appointments that are not in the Solicitor General's appendix); Recess Appointments, Washington Post, July 7, 1880, p. 1 (noting that President Hayes had made "quite a number of appointments" to pre-recess vacancies).
Then in 1940 Congress amended the Pay Act to authorize salary payments (with some exceptions) where (1) the "vacancy arose within thirty days prior to the termination of the session," (2) "at the termination of the session" a nomination was "pending," or (3) a nominee was "rejected by the Senate within thirty days prior to the termination of the session." Act of July 11, 54 Stat. 751 (codified, as amended, at 5 U.S.C. §5503 ). All three circumstances concern a vacancy that did not initially occur during a recess but happened to exist during that recess. By paying salaries to this kind of recess appointee, the 1940 Senate (and later Senates) in effect supported the President's interpretation of the Clause.
The upshot is that the President has consistently and frequently interpreted the Recess Appointments Clause to apply to vacancies that initially occur before, but continue to exist during, a recess of the Senate. The Senate as a body has not countered this practice for nearly three-quarters of a century, perhaps longer. See A. Amar, The Unwritten Constitution 576-577, n. 16 (2012) (for nearly 200 years "the overwhelming mass of actual practice" supports the President's interpretation); Mistretta v. United States, 488 U.S. 361 , 401 (1989) (a "200-year tradition" can " 'give meaning' to the Constitution" (quot-ing Youngstown, 343 U.S., at 610 (Frankfurter, J., concurring))). The tradition is long enough to entitle the practice "to great regard in determining the true construction" of the constitutional provision. The Pocket Veto Case, 279 U.S., at 690 . And we are reluctant to upset this traditional practice where doing so would seriously shrink the authority that Presidents have believed existed and have exercised for so long.
In light of some linguistic ambiguity, the basic purpose of the Clause, and the historical practice we have described, we conclude that the phrase "all vacancies" [*24] includes vacancies that come into existence while the Senate is in session.
The third question concerns the calculation of the length of the Senate's "recess." On December 17, 2011, the Senate by unanimous consent adopted a resolution to convene "pro forma session[s]" only, with "no business . . . transacted," on every Tuesday and Friday from December 20, 2011, through January 20, 2012. 2011 S. J. 923. At the end of each pro forma session, the Senate would "adjourn until" the following pro forma session. Ibid. During that period, the Senate convened and adjourned as agreed. It held pro forma sessions on December 20, 23, 27, and 30, and on January 3, 6, 10, 13, 17, and 20; and at the end of each pro forma session, it adjourned until the time and date of the next. Id., at 923-924; 158 Cong. Rec. S1-S11.
The President made the recess appointments before us on January 4, 2012, in between the January 3 and the January 6 pro forma sessions. We must determine the significance of these sessions-that is, whether, for purposes of the Clause, we should treat them as periods when the Senate was in session or as periods when it was in recess. If the former, the period between January 3 and January 6 was a 3-day recess, which is too short to trigger the President's recess-appointment power, see supra, at 19-21. If the latter, however, then the 3-day period was part of a much longer recess during which the President did have the power to make recess appointments, see ibid.
The Solicitor General argues that we must treat the pro forma sessions as periods of recess. He says that these "sessions" were sessions in name only because the Senate was in recess as a functional matter. The Senate, he contends, remained in a single, unbroken recess from January 3, when the second session of the 112th Congress began by operation of the Twentieth Amendment, until January 23, when the Senate reconvened to do regular business.
In our view, however, the pro forma sessions count as sessions, not as periods of recess. We hold that, for purposes of the Recess Appointments Clause, the Senate is in session when it says it is, provided that, under its own rules, it retains the capacity to transact Senate business. The Senate met that standard here.
The standard we apply is consistent with the Constitution's broad delegation of authority to the Senate to determine how and when to conduct its business. The Constitution explicitly empowers the Senate to "determine the Rules of its Proceedings." Art. I, §5, cl. 2 . And we have held that "all matters of method are open to the determination" of the Senate, as long as there is "a reasonable relation between the mode or method of proceeding established by the rule and the result which is sought to be attained" and the rule does not "ignore constitutional restraints or violate fundamental rights." United States v. Ballin, 144 U.S. 1 , 5 (1892).
In addition, the Constitution provides the Senate with extensive control over its schedule. There are only limited exceptions. See Amdt. 20, §2 (Congress must meet once a year on January 3, unless it specifies another day by law); Art. II, §3 (Senate must meet if the President calls it into special session); Art. I, §5, cl. 4 [*25] (neither House may adjourn for more than three days without consent of the other). See also Art. II, §3 ("[I]n Case of Disagreement between [the Houses], with Respect to the Time of Adjournment, [the President] may adjourn them to such Time as he shall think proper"). The Constitution thus gives the Senate wide latitude to determine whether and when to have a session, as well as how to conduct the session. This suggests that the Senate's determination about what constitutes a session should merit great respect.
Furthermore, this Court's precedents reflect the breadth of the power constitutionally delegated to the Senate. We generally take at face value the Senate's own report of its actions. When, for example, "the presiding officers" of the House and Senate sign an enrolled bill (and the President "approve[s]" it), "its authentication as a bill that has passed Congress should be deemed complete and unimpeachable." Marshall Field & Co. v. Clark, 143 U. S. 649 , 672 (1892). By the same principle, when the Journal of the Senate indicates that a quorum was present, under a valid Senate rule, at the time the Senate passed a bill, we will not consider an argument that a quorum was not, in fact, present. Ballin, supra, at 9 . The Constitution requires the Senate to keep its Journal, Art. I, §5, cl. 3 ("Each House shall keep a Journal of its proceedings . . ."), and "if reference may be had to" it, "it must be assumed to speak the truth," Ballin, supra, at 4 .
For these reasons, we conclude that we must give great weight to the Senate's own determination of when it is and when it is not in session. But our deference to the Senate cannot be absolute. When the Senate is without the capacity to act, under its own rules, it is not in session even if it so declares. See Tr. of Oral Arg. 69 (acknowledgment by counsel for amici Senators that if the Senate had left the Capitol and "effectively given up . . . the business of legislating" then it might be in recess, even if it said it was not). In that circumstance, the Senate is not simply unlikely or unwilling to act upon nominations of the President. It is unable to do so. The purpose of the Clause is to ensure the continued functioning of the Federal Government while the Senate is unavailable. See supra, at 5-6. This purpose would count for little were we to treat the Senate as though it were in session even when it lacks the ability to provide its "advice and consent." Art. II, §2, cl. 2 . Accordingly, we conclude that when the Senate declares that it is in session and possesses the capacity, under its own rules, to conduct business, it is in session for purposes of the Clause.
Applying this standard, we find that the pro forma sessions were sessions for purposes of the Clause. First, the Senate said it was in session. The Journal of the Senate and the Congressional Record indicate that the Senate convened for a series of twice-weekly "sessions" from December 20 through January 20. 2011 S. J. 923-924; 158 Cong. Rec. S1-S11. (The Journal of the Senate for 2012 has not yet been published.) And these reports of the Senate "must be assumed to speak the truth." Ballin, supra, at 4 .
Second, the Senate's rules make clear that during its pro forma [*26] sessions, despite its resolution that it would conduct no business, the Senate retained the power to conduct business. During any pro forma session, the Senate could have conducted business simply by passing a unanimous consent agreement. See Riddick's 1313. The Senate in fact conducts much of its business through unanimous consent. Id., at 1311-1312. Senate rules presume that a quorum is present unless a present Senator questions it. Id., at 1041-1042. And when the Senate has a quorum, an agreement is unanimously passed if, upon its proposal, no present Senator objects. Id., at 1329-1330. It is consequently unsurprising that the Senate has enacted legislation during pro forma sessions even when it has said that no business will be transacted. Indeed, the Senate passed a bill by unanimous consent during the second pro forma session after its December 17 adjournment. 2011 S. J. 924. And that bill quickly became law. Pub. L. 112-78 , 125 Stat. 1280 .
By way of contrast, we do not see how the Senate could conduct business during a recess. It could terminate the recess and then, when in session, pass a bill. But in that case, of course, the Senate would no longer be in recess. It would be in session. And that is the crucial point. Senate rules make clear that, once in session, the Senate can act even if it has earlier said that it would not.
The Solicitor General argues that more is required. He contends that what counts is not the Senate's capacity to conduct business but what the Senate actually does (or here, did) during its pro forma sessions. And he looks for support to the functional definition of "recess" set forth in the 1905 Senate Report discussed above. See supra, at 14. That Report describes a "recess" of the Senate as
"the period of time . . . when its members owe no duty of attendance; when its Chamber is empty; when, because of its absence, it can not receive communications from the President or participate as a body in making appointments." 1905 Senate Report, at 2.
Even were we, for argument's sake, to accept all of these criteria as authoritative, they would here be met. Taking the last criterion first, could the Senate, during its pro forma sessions, "participate as a body in making appointments"? It could. It could confirm nominees by unanimous consent, just as it passed the bill mentioned above. See Riddick's 1313.
Could the Senate "receive communications from the President"? It could. The Congressional Record indicates that the Senate "received" a message from the President on January 12, during a 3-day adjournment between two pro forma sessions. See 158 Cong. Rec. S37 (Jan. 23, 2012). If the Senate could receive Presidential messages between two pro forma sessions, it could receive them during a pro forma session.
Was the Senate's Chamber "empty"? It was not. By its official rules, the Senate operates under the presumption that a quorum is present until a present Senator suggests the absence of a quorum, Riddick's 1041-1042, and nothing in the Journal of the Senate or the Congressional Record reflects any such suggestion.
Did Senators "owe [a] duty of attendance"? They did. The Senate's rules dictate that Senators [*27] are under a duty to attend every session. See Riddick's 214; Standing Rule of the Senate VI(2), S. Doc. No. 112-1, p. 5 (2011) ("No Senator shall absent himself from the service of the Senate without leave"). Nothing excused the Senators from this duty during the Senate's pro forma sessions. If any present Senator had raised a question as to the presence of a quorum, and by roll call it had become clear that a quorum was missing, the Senators in attendance could have directed the Sergeant at Arms to bring in the missing Senators. Rule VI(4).
The Solicitor General asks us to engage in a more realistic appraisal of what the Senate actually did. He argues that, during the relevant pro forma sessions, business was not in fact conducted; messages from the President could not be received in any meaningful way because they could not be placed before the Senate; the Senate Chamber was, according to C-SPAN coverage, almost empty; and in practice attendance was not required. See Brief for Petitioner 48-49, 54-55.
We do not believe, however, that engaging in the kind of factual appraisal that the Solicitor General suggests is either legally or practically appropriate. From a legal perspective, this approach would run contrary to prece-dent instructing us to "respect . . . coequal and independent departments" by, for example, taking the Senate's report of its official action at its word. Field, 143 U.S., at 672 ; see Ballin, 144 U.S., at 4 . From a practical perspective, judges cannot easily determine such matters as who is, and who is not, in fact present on the floor during a particular Senate session. Judicial efforts to engage in these kinds of inquiries would risk undue judicial interference with the functioning of the Legislative Branch.
Finally, the Solicitor General warns that our holding may " 'disrup[t] the proper balance between the coordinate branches by preventing the Executive Branch from accomplishing its constitutionally assigned functions.' " Brief for Petitioner 64 (quoting Morrison v. Olson, 487 U.S. 654 , 695 (1988); alteration in original). We do not see, however, how our holding could significantly alter the constitutional balance. Most appointments are not controversial and do not produce friction between the branches. Where political controversy is serious, the Senate unquestionably has other methods of preventing recess appointments. As the Solicitor General concedes, the Senate could preclude the President from making recess appointments by holding a series of twice-a-week ordinary (not pro forma) sessions. And the nature of the business conducted at those ordinary sessions-whether, for example, Senators must vote on nominations, or may return to their home States to meet with their constituents-is a matter for the Senate to decide. The Constitution also gives the President (if he has enough allies in Congress) a way to force a recess. Art. II, §3 ("[I]n Case of Disagreement between [the Houses], with Respect to the Time of Adjournment, [the President] may adjourn them to such Time as he shall think proper"). Moreover, the President and Senators engage with each other in many different ways [*28] and have a variety of methods of encouraging each other to accept their points of view.
Regardless, the Recess Appointments Clause is not designed to overcome serious institutional friction. It simply provides a subsidiary method for appointing officials when the Senate is away during a recess. Here, as in other contexts, friction between the branches is an inevitable consequence of our constitutional structure. See Myers, 272 U.S., at 293 (Brandeis, J., dissenting). That structure foresees resolution not only through judicial interpretation and compromise among the branches but also by the ballot box.
The Recess Appointments Clause responds to a structural difference between the Executive and Legislative Branches: The Executive Branch is perpetually in operation, while the Legislature only acts in intervals separated by recesses. The purpose of the Clause is to allow the Executive to continue operating while the Senate is unavailable. We believe that the Clause's text, standing alone, is ambiguous. It does not resolve whether the President may make appointments during intra-session recesses, or whether he may fill pre-recess vacancies. But the broader reading better serves the Clause's structural function. Moreover, that broader reading is reinforced by centuries of history, which we are hesitant to disturb. We thus hold that the Constitution empowers the President to fill any existing vacancy during any recess-intra-session or inter-session-of sufficient length.
JUSTICE SCALIA would render illegitimate thousands of recess appointments reaching all the way back to the founding era. More than that: Calling the Clause an "anachronism," he would basically read it out of the Constitution. Post, at 12. He performs this act of judicial excision in the name of liberty. We fail to see how excising the Recess Appointments Clause preserves freedom. In fact, Alexander Hamilton observed in the very first Feder-alist Paper that "the vigour of government is essential to the security of liberty." The Federalist No. 1, at 5. And the Framers included the Recess Appointments Clause to preserve the "vigour of government" at times when an important organ of Government, the United States Senate, is in recess. JUSTICE SCALIA's interpretation of the Clause would defeat the power of the Clause to achieve that objective.
The foregoing discussion should refute JUSTICE SCALIA's claim that we have "embrace(d)" an "adverse-possession theory of executive power." Post, at 48. Instead, as in all cases, we interpret the Constitution in light of its text, purposes, and "our whole experience" as a Nation. Missouri v. Holland, 252 U.S. 416 , 433 (1920). And we look to the actual practice of Government to inform our interpretation.
Given our answer to the last question before us, we conclude that the Recess Appointments Clause does not give the President the constitutional authority to make the appointments here at issue. Because the Court of Appeals reached the same ultimate conclusion (though for reasons we reject), its judgment is affirmed.
It is so ordered.
High court to hear Obama recess appointments case